30 Carl Fred. Kolderup. [Nr. 18 



unlike to be put into one group. Some of them have the chemical com- 

 position of the granites containing more than 66 % Si 2 , others are 

 quartzdiorites. In a geological point of view they form one series, 

 ftut petrografi cally they must be separated in two divisions. I 

 fully agree with Hatch (Textbook of petrology) who says: „Ac- 

 cording to Lindgeen the percentage of silica which rises as 

 high as 73 per cent, may fall as low as 59 per cent, and he makes 

 the distinction from the more acid members of the diorite family 

 depend on the orthoclase. But most, if not all, diorites contain 

 some small quantity of orthoclase, and it is better for practical 

 purpose to make the percentage of the silica the means of dividing 

 the granodiorites from the diorites, and to call no rock with less 

 than 66 per cent of silica a granodiorite." As already mentioned 

 the white granites of Svanø, Sogneskollen and Bremangerland must 

 be regarded as granodiorites; they will thus be the most acid types 

 of granodiorites, hitherto analysed. 



When we put these analyses into the series mentioned above, 

 we shall get a good general view of the chemical composition of 

 the granodiorite sub family (the analyses page 25). The different 

 percentages vary within the following borders: 



Si 2 72.13—66.40 



Al 2 3 14.95-17.13 



Fe a 3 + Fe 0... 2.00— 3.97 



Mg 0.09— 1.74 



Ca 1.71— 4.53 



Na 2 5.99— 3.40 



K 2 1.71— 2.65 (3.66) 



The percentage of K 2 rises in one case to 3.66 % but as the 

 percentage of the K 2 of this rock is higher than of Na 2 0, I am 

 not quite certain whether it would not be better to call the rock 

 -an adamellite. 



